The modern data science landscape is characterized by an unprecedented diversity of data, spanning formats, sources, and domains. With the proliferation of sensor data, unstructured text, multimedia, and structured records from disparate platforms, artificial intelligence (AI) and machine learning (ML) systems must rise to the challenge of robustly and securely learning from these heterogeneous data sources. Applications across industries such as healthcare, finance, cybersecurity, and IoT increasingly rely on AI to derive actionable insights from such complex data ecosystems. However, achieving this requires overcoming the following critical challenges:
1. Variability in data formats and representations.
2. Missing, incomplete, or imprecise data.
3. Privacy, security, and adversarial robustness concerns.
4. Real-time learning requirements from multi-source data streams.
This Research Topic invites innovative solutions and interdisciplinary approaches to address these challenges, with a particular emphasis on developing resilient, secure, and interpretable AI/ML systems capable of seamlessly learning from and integrating diverse data modalities.
The topics of interests include, but are not limited to:
Data Representation and Integration for Heterogeneous Data:
• Novel representation techniques to bridge semantic gaps across data sources.
• Frameworks for efficient integration of structured, semi-structured, and unstructured data.
Robust and Secure AI/ML Algorithms:
• Development of algorithms for multi-source data fusion and analysis.
• Techniques for privacy-preserving learning, including federated learning and adversarial robustness.
Real-time and Multi-modal Learning:
• Methods for fusing data from multiple modalities such as text, vision, audio, and sensors.
• Real-time learning from streaming heterogeneous data in dynamic environments.
Explainable and Interpretable AI:
• Transparent AI systems for multi-source data analysis and decision-making.
• Improving interpretability in complex, multi-modal models.
Domain-specific Applications:
• Case studies in healthcare, finance, cybersecurity, IoT, and other industries.
• Addressing domain-specific challenges using heterogeneous data solutions.
Keywords: Heterogeneous Data Integration, Robust Machine Learning, Privacy-Preserving AI, Multi-modal Data Fusion, Explainable AI
Important note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
